The Food Lab, Drinks Edition: Is Mexican Coke Better?

Everyone in the know knows that sugar-sweetened Mexican Coke is superior to corn-syrup-sweetened American Coke, right?

Mexican and American Coke cans chilling in ice.

Serious Eats / J. Kenji López-Alt

Let's start with a little quote from The Philosophy of Andy Warhol:

What's great about this country is that America started the tradition where the richest consumers buy essentially the same things as the poorest. You can be watching TV and see Coca-Cola, and you know that the President drinks Coke, Liz Taylor drinks Coke, and just think, you can drink Coke, too. A Coke is a Coke and no amount of money can get you a better Coke than the one the bum on the corner is drinking. All the Cokes are the same and all the Cokes are good. Liz Taylor knows it, the President knows it, the bum knows it, and you know it.

And it's all true enough, right? Apparently not. If there's one thing this country is really great at, it's coming up with clever new ways to take what is a completely normal product, apply a bit of subtle psychological manipulation, convince people that it's something special, and sell it at a jacked up price.

I'm talking here about Mexican Coke, and I do so not without a hint of irony, because I myself am a firm believer in its superiority over regular old American Coke. I mean, how could it not be better? Real sugar instead of corn syrup. Glass bottle instead of aluminum or plastic. The cachet of seeing the words refresco and no retornable printed instead of plain old pedestrian "refreshing."

It's so much better, in fact, that I go out of my way to seek it out. I keep a little black notebook of the elusive locations selling it. Everywhere from the Costco on 117th and Pleasant to the bodega around the corner from the office, to the restaurants that are hip to its superiority (and charge an arm and a leg for it).

You want Mexican Coke? I can get you a Mexican Coke. There are ways, believe me. Hell, I could get you a Mexican Coke by 3 o'clock this afternoon, with a bottle opener.

Ten plastic cups containing Coke prepared for a taste test.

Serious Eats / J. Kenji López-Alt

But here's the thing. More than once in the past, I've discovered that the brain has a powerful effect on the taste buds. Free-range eggs taste better? Nope. Darker-colored eggs taste better. Is New York pizza better when made with New York tap water? Nope. At least my panel of experts couldn't tell the difference. I've done tests where I've fed an entire room full of people two batches of identical carrots, labeling one as organic and the other as conventional. Unsurprisingly, they unanimously pick the carrots labeled organic as superior in flavor every single time, even when they are two halves of the same carrot.

Is it possible, however unlikely, that somehow we—the cult of Mexican Coke lovers—are all being hoodwinked? Does Mexican Coke really taste better?

We're gonna find out.

Behind The Bottle

First off, before we even get to the tasting, let's examine the differences between regular old American Coke and Mexican Coke.

  • Mexican Coke contains: Carbonated water, sugar, caramel color, phosphoric acid, natural flavors, caffeine.
  • American Coke contains: Carbonated water, high fructose corn syrup, caramel color, phosphoric acid, natural flavors, caffeine.

Since 1980, American Coke has been formulated with High-Fructose Corn Syrup (HFCS) instead of Sucrose (that is, table sugar). Now I'm no nutritionist or dietitian so I'm not going to weigh in on the whole HFCS vs. real sugar thing from a health perspective except to say that as someone with a very basic understanding of chemistry, I don't worry myself too much about the health implications of HFCS vs. sugar. After all, they are nearly chemically identical. Sucrose (or saccharose) is a disaccharide made of a single fructose molecule attached to a single glucose molecule.

HFCS is a mixture of dissociated glucose and fructose molecules. The only real difference between the two is the ratio of fructose to glucose, and the way in which the molecules are connected with each other. Even sucrose breaks down into fructose and glucose in your body. I feel pretty safe drinking either version (at least in moderation).

At first glance, the labels on both bottles appear to be pretty much identical, except for their sweetener. But take a closer look at the nutrition information and you'll find that for the same 355 mL serving, American Coke has 140 calories and 45 mg of sodium, while the Mexican version has 150 calories and 85 mg of sodium. How do you explain this?

I called up the Coca-Cola company to find out and talked a bit with a very pleasant lady named Annette. After a bit of idle chit-chat on the state of cola we got down to brass tacks. Or at least, I tried to.

"The differences all come down to formulation," she'd tell me.

Ok, thanks. But Annette, I'm wondering, exactly what kind of differences?

Well, depending on what part of the world you go to, Coke is made with a slightly different formula."

I understand that, but is it to appease local palates? Is it because of ingredient availability? What's the market research? Dear God, Annette, where are the data?!?

"Oh, I see what you're asking, sir. Well, the truth is, because Coke is made with a different formula in each part of the world, the formulation is not the same from country to country."

Yep, the old runaround. It all reminded me very much of my early dating career: feigned interest, pleasant enough conversation, a general airing of grievances but ultimately, no real return on my time investment. And for the record, I have absolutely no information on whether or not Coke is planning on releasing a competitor to the sugar-sweetened Pepsi Throwback.

Without a straight-up answer from Coke, I can only speculate as to the differences. Since there is exactly the same amount of fat, protein, and carbohydrates by weight, the difference in calorie count must simply be in the rounding (companies are required to report calories in multiples of ten). There are 3.8 calories in a gram of sugar, giving us an actual calorie count of 148.2 per 355 ml bottle. Somewhere in between the reported 140 and 150.

As for sodium, my guess would be that the Mexican and American bottlers use different carbonation methods. More sodium bicarbonate (an ingredient used to give club soda its bubbles) would lead to a higher sodium content.

Whether all of this affects the flavor of the two products is the real question.

The Tasting


Serious Eats / J. Kenji López-Alt

For the purposes of my taste test there were a couple of criteria I had to set up first:

  • Mexican Coke would come in bottles, American Coke would come in cans. Of the packaging widely available in America (plastic or aluminum), aluminum is less reactive, less porous, more opaque, has a longer shelf life, and is thus more likely to give me a product that simply tastes more like it should.
  • All Coke must be served ice cold. Bottles and cans would be stored in the fridge then placed in an ice water bath for at least one hour before tasting.
  • All Coke must be as fresh as possible. According to Annette, canned Coke and Mexican glass-bottled Coke both have a shelf life of nine months (plastic bottle Coke, on the other hand, starts losing bubbles after a mere 10 weeks). I managed to find cases of Mexican Coke and American Coke with expiration dates within a week of each other next April.

Now, a lazy researcher could crack open a couple cans and bottles, invite some friends over, ask for some opinions, and be done with the whole thing, and there's no shortage of "taste tests" on the internet of this sort. But we all know that those tests aren't really valid, right? I mean, double-blind, good science, and all that?

What we'd be doing is subjecting our tasters to an entire battery of tests. See, I'm not so convinced that a lot of what's going on in your mouth isn't based solely on the packaging or presentation of the Coke product. Indeed, perhaps the packaging and presentation are even more important than the flavor itself. If the Coke 2 debacle of the 1980s* taught us anything, it's that when it comes down to it, people care far more about branding than actual flavor. Would this be the case with Mexican Coke as well?

*An infamous marketing play gone bad in which the Coca-Cola company carried out a series of blind taste tests between its classic Coke, a newly reformulated "New Coke," and Pepsi. In the blind taste tests, tasters overwhelmingly preferred the flavor of New Coke to either Coke or Pepsi. Yet when it was released later on as "New Coke," the Coca-Cola loyalists flipped out. Eventually, the company decided to re-release the old-fashioned Coke as "Coca-Cola Classic" (branding they still use to this day) and finally re-branded the new Coca-Cola as "Coke 2" in 1992. It never sold well (despite people preferring its flavor), and eventually slipped quietly into the night in the early 2000's.

Here's what I tested in my first round. All tests were carried out completely blind. Tasters were brought one at a time to taste and did not discuss their answers with either myself nor any of the other tasters until all responses were completely collected. For each taster, tests were administered in a completely random order (both in terms of test order and sample order), and fresh bottles and cans were opened for each taster. In cases where liquid had to be poured from one vessel to another, the utmost care was taken to ensure a minimal loss of carbonation. Tasters were asked to pick their favorite from within each sample set of two.

  • Test 1: Mexican Coke in glass bottle vs. American Coke in a can.
  • Test 2: Mexican Coke in a cup with ice vs. American Coke in a cup with ice.
  • Test 3: Mexican Coke in a cup with no ice vs. American Coke in a cup with no ice.
  • Test 4: Mexican Coke in a can vs. American Coke in a can.
  • Test 5: Mexican Coke in a can vs. American Coke in a glass bottle.
  • Test 6: Mexican Coke in a glass bottle vs. American Coke in a glass bottle.
  • Test 7: American Coke in a can vs. American Coke in a glass bottle.

With this battery of tests—which pitted the most important permutations of American/Mexican and can/bottle/cup against one another—I was fairly confident that I should be able to tease out whether or not tasters could a) really taste a difference between the two products, b) whether they preferred drinking from a can or a bottle, and c) whether the difference a can or bottle makes is great enough that it trumps any perceived flavor differences.

The Tasters and the Feelers


Serious Eats / J. Kenji López-Alt

The spread of results I got from this initial testing was surprising to say the least, and answered one thing for sure: There is a perceivable difference in the flavor between Mexican and American Coke, despite the best efforts of the Coca-Cola company to convince us otherwise.

The first analysis I made was to tally up the scores between every test in which tasters had a choice between Mexican and American Coke (that is, tests 1 through 6). From within this set of tests, there was an overwhelming preference for American Coke over Mexican Coke. The average taster picked regular coke two to one over Mexican coke!

So that settles it. America reigns supreme in the Coke flavor wars, right? Not so fast. Looking closer, we see something even more interesting: Half of the tasters seemed to have no real preference between American and Mexican Coke, while the other half of the tasters unanimously chose American Coke as their favorite for nearly every test, regardless of the vessel it was served in. We'll call these folks the Tasters—the ones who let their tongues and noses do all the deciding.

The Tasters pick out American Coke as superior to Mexican Coke a full 7 times out of 8.

When you take the Tasters out of the pool in order to determine what the other half are basing their tasting decision on, everything becomes clear: The other half of the tasters unanimously picked Coke served out of a glass bottle as their favorite for nearly each and every test, regardless of whether the liquid in there was Mexican or American Coke. We'll call these folks the Feelers—the ones who care more about the tactile sense of the bottle against their lips or in their hands than the minor differences in flavor or aroma that the product inside may have.


Serious Eats / J. Kenji López-Alt

So just to sum up here:

  • People prefer American Coke to Mexican Coke from a pure flavor and aroma standpoint.
  • People prefer glass bottles to aluminum cans from a purely tactile standpoint.

Interesting, right?

But there's still a nagging question on my mind. Why don't people freak out about American Coke sold in glass bottles? I mean, it's available, and it theoretically should make both the Tasters and the Feelers happy, providing the ultimate Coke experience, right?

I did one more round of testing with a fresh batch of tasters to get to the bottom of it.

The Mexico Boosters

This time, there would be no more trickery. Ok, I lie. A little bit more trickery. My goal was to see if in a non-blind taste test—that is, one in which the tasters are specifically told that what they are drinking is Mexican Coke or American Coke (regardless of if it really is or not), would they be consistent in their choices?

That is, could the knowledge that a given batch of Coke is Mexican affect tasters' perception of it?

For this set of tests, I had tasters try Mexican and American Coke out of glass bottles and cans. For half of the tests, tasters were told the truth: When I said the Mexican Coke was in the glass bottle and the American Coke was in the can, it really was. For the other half of the tests, I told a lie. Both the Coke in the can and the bottle were from the exact same source.

Can you guess what happened?

Exactly. Regardless of what was actually in the serving containers, people stuck by their original choice. Those who preferred what really was the Mexican Coke the first time (we'll call these guys the Mexico Boosters) unanimously picked the Coke that I told them was the Mexican Coke the second time, whether it really was or was not. Even when the containers were completely removed from the test and the Coke was served in plastic cups, the Coke labeled as Mexican was picked by the Mexico Boosters every time.

Of course, that's not to say that everyone was a Mexico Booster. Some folks knowingly picked American Coke (though they too consistently picked the Coke labeled American, regardless of whether it really was or what container it was served in).

What Does It Mean?


Serious Eats / J. Kenji López-Alt

There are a number of pretty clear conclusions that can be drawn from these tests. To put it simply, when it comes to taste, there's this simple relationship: Boosterism > Tasting = Feeling, meaning that while there are an equal number of people who are affected by the flavor of Coke as there are affected by the feel of the container, both of these groups are eclipsed once you add in the knowledge of the product's provenance. Those folks who prefer Mexican Coke (like myself), really just like the idea of Mexican Coke—whether it's because they think real sugar is tastier/healthier than corn syrup, whether it's because Mexican Coke is more expensive and harder to find, thus more valuable, whether it's because of its exoticism, whatever the reason—strip away the Mexicanness of it, and suddenly it's a lot less appealing.

This is not all that surprising to me, given some pretty similar results in taste tests past. What was surprising was that after the Mexicanity of the Coke was removed, people actually preferred the flavor of American Coke.

So, Coca-Cola Company, here is what you need to do to provide your valued customers with the ultimate Coke experience: Bottle American HFCS-sweetened Coke in Mexican Coke bottles, and just tell everyone it comes from Mexico.

There are many questions left unanswered here: Does glass insulate better than aluminum, and does that have an effect? How about plastic? What about storage conditions—can light affect flavor? What about quantity? Our tasters were taking sips. Would their opinions change if they had to drink an entire portion? And for some of us, more importantly, what about Diet Coke vs. Coke Light?

The universe never ceases to amaze with the number of questions it can throw your way. Now, I've gotta go running around the block a few times to work off the massive caffeine shakes I've built up.

September 2011